Antares raises $470M to deploy small modular nuclear reactors on US military bases

Antares Raises $470M to Build Nuclear Reactors for the Pentagon, Not the Power Grid

Antares Nuclear closed a $470 million Series C on Monday to build small modular reactors for U.S. military bases. The deeper signal inside that raise: advanced fission's first viable revenue path runs through the Department of Defense, not the commercial grid. Which reactor designs get proven on military installations will shape what reaches AI data centers when the power crunch intensifies.

The round, led by Paradigm and Caffeinated Capital with participation from Industrious Ventures, Point72 Ventures, and Shine Capital, included $370 million in equity and $100 million in debt 1. It brings Antares' total funding to $604 million, per TechCrunch's analysis of PitchBook data 1.

A reactor that already works

The Mark-0 demonstration reactor achieved first criticality at Idaho National Laboratory on June 4 1. The reactor runs on TRISO fuel, uranium particles sealed in carbon and ceramic layers that resist meltdown at high temperatures. Its SMR is designed to generate 100 kilowatts to 1 megawatt, sufficient for about 750 homes 1.

Antares was selected as one of three finalists in the Defense Department's Advanced Nuclear Power for Installations program, which will evaluate small modular reactors at Air Force installations in Colorado and Montana 1. The company plans to bring its first electricity-producing reactor online next year and deploy at military sites by 2028 1.

Why the military, not the grid

Lazard estimates new SMRs will cost roughly $214 per megawatt hour 1. At that price, commercial deployment is economically impossible for utilities today. The Department of Defense operates under different incentives. It values energy resilience and operational independence over cost per kilowatt hour. The DOE describes microreactors as factory-fabricated, transportable by truck or aircraft, and capable of operating for up to 10 years without refueling 2, features that map directly onto military requirements for remote and forward-deployed installations.

Lazard frames the broader moment as a "speed-to-power era" where demand is outpacing supply across every generation technology 3. Renewables remain the cheapest new capacity, but the analysis stresses that meeting unprecedented demand will require a diverse generation fleet.

The funding surge

Antares is not raising alone. X-energy raised $1 billion through an IPO in April, while Radiant Energy, Standard Nuclear, and Last Energy each closed nine-figure rounds since December 1. Antares' previous round, a $96 million Series B, also closed in December 1. The acceleration tracks the AI data center buildout and broader electrification, both pushing electricity demand past what existing infrastructure can supply.

What this means for commercial AI

Defense procurement is functioning as a selection filter. The reactor designs that win DoD contracts in 2028 will accumulate manufacturing experience, regulatory milestones, and operational data. Those that don't will reach commercial markets later, with thinner track records, at a moment when AI infrastructure operators are already short on time and power.

TechCrunch notes that mass manufacturing benefits typically take at least a decade to materialize, and no SMR startup has reached that stage yet 1. Military contracts buy the operational years and supply-chain maturity that startups need to eventually reach commercial cost competitiveness. By the time SMRs are cheap enough for utilities, the companies that served the Pentagon first will have the deepest manufacturing curves and the strongest reliability records.

Antares' $470 million is a wager that the path to commercial nuclear runs through a military base in Colorado or Montana. The investors backing that bet are staking their position on a simple premise: which reactor gets proven first determines who sells to the grid second.